BS-Net: Learning COVID-19 pneumonia severity on a large chest X-ray dataset

Autor: Sergio Benini, Davide Farina, Nicola Adami, Filippo Vaccher, Alberto Signoroni, Andrea Borghesi, Marco Ravanelli, Mattia Savardi, Paolo Gibellini, Riccardo Leonardi, Roberto Maroldi
Rok vydání: 2021
Předmět:
FOS: Computer and information sciences
Computer Science - Machine Learning
J.3
Source code
Computer science
Computer Vision and Pattern Recognition (cs.CV)
Computer Science - Computer Vision and Pattern Recognition
computer.software_genre
Convolutional neural network
COVID-19 severity assessment
Machine Learning (cs.LG)
030218 nuclear medicine & medical imaging
0302 clinical medicine
semi-quantitative rating
Segmentation
I.2.10
I.5
I.4
End-to-end learning
media_common
Radiological and Ultrasound Technology
Convolutional Neural Networks
Image and Video Processing (eess.IV)
Chest X-Rays
Computer Graphics and Computer-Aided Design
Radiography
Thoracic
Computer Vision and Pattern Recognition
media_common.quotation_subject
Health Informatics
Machine learning
Article
03 medical and health sciences
Consistency (database systems)
Deep Learning
Robustness (computer science)
FOS: Electrical engineering
electronic engineering
information engineering
Humans
Radiology
Nuclear Medicine and imaging
ComputingMethodologies_COMPUTERGRAPHICS
SARS-CoV-2
business.industry
X-Rays
Deep learning
Supervised learning
COVID-19
Gold standard (test)
Electrical Engineering and Systems Science - Image and Video Processing
68T45
Chest X-rays
Convolutional neural networks
Semi-quantitative rating
Artificial intelligence
business
computer
030217 neurology & neurosurgery
Zdroj: Medical Image Analysis
ISSN: 1361-8415
DOI: 10.1016/j.media.2021.102046
Popis: Graphical abstract
In this work we design an end-to-end deep learning architecture for predicting, on Chest X-rays images (CXR), a multi-regional score conveying the degree of lung compromise in COVID-19 patients. Such semi-quantitative scoring system, namely Brixia score, is applied in serial monitoring of such patients, showing significant prognostic value, in one of the hospitals that experienced one of the highest pandemic peaks in Italy. To solve such a challenging visual task, we adopt a weakly supervised learning strategy structured to handle different tasks (segmentation, spatial alignment, and score estimation) trained with a “from-the-part-to-the-whole” procedure involving different datasets. In particular, we exploit a clinical dataset of almost 5,000 CXR annotated images collected in the same hospital. Our BS-Net demonstrates self-attentive behavior and a high degree of accuracy in all processing stages. Through inter-rater agreement tests and a gold standard comparison, we show that our solution outperforms single human annotators in rating accuracy and consistency, thus supporting the possibility of using this tool in contexts of computer-assisted monitoring. Highly resolved (super-pixel level) explainability maps are also generated, with an original technique, to visually help the understanding of the network activity on the lung areas. We also consider other scores proposed in literature and provide a comparison with a recently proposed non-specific approach. We eventually test the performance robustness of our model on an assorted public COVID-19 dataset, for which we also provide Brixia score annotations, observing good direct generalization and fine-tuning capabilities that highlight the portability of BS-Net in other clinical settings. The CXR dataset along with the source code and the trained model are publicly released for research purposes.
Databáze: OpenAIRE
Externí odkaz: